When is a QSO Complete ?

Once you start to decode EME signals, it is natural to want to have a go, but before you do this, make sure that you understand the procedure used in an EME QSO.
This has evolved over many years so as to optimize the chance of completing the QSO, and it is surprising how many stations fail to follow it closely.
Joe Taylor has written a concise Help entry in the WSJT program - click on Help, then What Message to Send? Study this carefully and make sure that you understand it fully - this will take longer than you think!

Stick to the basic procedure exactly until the QSO is legally complete, only then add some chit-chat if you want.

Formal Requirements of a QSO

The formal requirements for completion of a QSO are:

  1. Both stations must copy/decode BOTH callsigns,
  2. Both stations must copy/decode a signal report, normally O, but traditionally sent as OOO,
  1. Both stations must copy/decode an acknowledgement that the report they each sent was received, normally R, traditionally sent as RRR.

It is not permitted to send the O report until you have copied both callsigns.
This may seem to be an unnecessary restriction, but it avoids ambiguity under marginal conditions.
If you receive an OOO report, you can be sure that the other station has copied both callsigns, and you need not send callsigns any longer.

In a typical sched situation, both stations start by sending callsigns.
The grid locator is attached by the program, and is useful for locating stations when collecting gridsquares, but the grid is not formally required for the QSO to be complete.

When either station decodes both callsigns, he responds by attaching his report OOO to the callsigns he continues to send.
If you receive callsigns + OOO, then you know that both stations have now copied both callsigns, and that you have received a report.
So you respond by acknowledging the other station's report with R, and also sending your report O to him, put together as RO.
Once he receives your RO, the only thing missing is his acknowledgement that he has received your O (part of the RO).
He therefore responds with RRR, and as soon as you receive that, the QSO is legally complete.

It all sounds more complicated than it really is, but you do need to grasp it thoroughly.

Notice that when you receive his RRR, you know that the QSO is complete, but the other station does not know this until you tell him so by sending 73.
A lot of twaddle has been written by experienced operators to the effect that if both stations don't know that the QSO is complete then it isn't. This is obviously rubbish!
The QSO is complete at the time that one station (suppose it to be you) receives RRR.
Now suppose that before you can send him 73 to tell him of the completion, your transmitter blows up (this really happened to me once), the mains power supply fails, your computer crashes, or the moon sets.
There is no way he can know that the QSO is complete until you tell him (by email, logger, post, or whatever). But the QSO was still legally complete.

There is no way out of this bind. Suppose the rules were changed to require your 73 to be received by the other station before the QSO could be deemed complete.
Now suppose his transmitter blows up after receiving your 73, but before he can tell you this .....

Look at the list of messages provided in JT65B. Messages 2 to 4 are each sent by only ONE of the two stations, and follow in a natural sequence.
Message 1 may well be sent by both stations in a sched situation (but not necessarily), and it is polite for both stations to send 73, but you should never respond to RO with RO, or RRR with RRR.
It simply isn't necessary, and the wasted time may make the QSO impossible to complete.
That said, if you understand the requirements then you can easily improvise if the other station gets it wrong.

For example, if you send him RO and he responds with RO instead of RRR, then O and R have been decoded in both directions, and that is all that is required, so the QSO is complete.
You can properly respond with 73, rather than send RRR.

Two small final things. Some computers are unable to complete the decode before the minute rolls over and you start transmitting.
No matter, you can change the message in mid transmission without damaging anything. There is no need to wait for another sequence.

And do make use of the waterfall display for decoding RO, RRR, and 73. This way you can be ready for the next step long before the computer does the decode.
Sometimes the computer will even fail to decode one of these shorthand messages if an interfering signal is close by.
It is often better to turn off the shorthand decodes completely, to avoid a message being mistakenly decoded as RO, RRR or 73 due to another sync tone just at the shorthand spacing.

Example

I once had the following JT65B EME exchange on 144MHz.
I have changed the callsign of the other station to avoid offence.
Each line is tagged at the start with the callsign of the transmitting station.

VK2KU : CQ VK2KU QF55
EU1AA : VK2KU EU1AA JN99 OOO
VK2KU : EU1AA VK2KU QF55 OOO
EU1AA : RO
VK2KU : RRR
EU1AA : RRR
VK2KU : 73
EU1AA: 73

EU1AA made an error in line 2 by sending OOO before he had received both callsigns - I had not even transmitted his callsign yet!
I therefore ignored the OOO, but since I had received both callsigns, I responded with callsigns + OOO.
His response of RO and my following response of RRR were correct. When he copied my RRR, the QSO was legally complete.
EU1AA made a second error by sending RRR when he could have gone straight to 73, but the QSO then proceeded to a normal completion.

The correct (and expected) sequence of messages for that QSO would have been:

VK2KU : CQ VK2KU QF55
EU1AA : VK2KU EU1AA JN99
VK2KU : EU1AA VK2KU QF55 OOO
EU1AA : RO
VK2KU : RRR
EU1AA: 73
VK2KU : 73

 


VK2KU - 13 January 2011